Dark Model Adaptation: Semantic Image Segmentation from Daytime to Nighttime

TL;DR

This paper introduces a novel approach for adapting semantic image segmentation models from daytime to nighttime scenes using twilight as a bridge, reducing the need for additional human annotations.

Contribution

The paper proposes a new method for progressive model adaptation from daytime to nighttime scenes and provides a large-scale dataset with annotations for evaluation.

Findings

Effective adaptation from daytime to nighttime scenes

Reduces need for human annotation in nighttime segmentation

Demonstrates superior performance on the new dataset

Abstract

This work addresses the problem of semantic image segmentation of nighttime scenes. Although considerable progress has been made in semantic image segmentation, it is mainly related to daytime scenarios. This paper proposes a novel method to progressive adapt the semantic models trained on daytime scenes, along with large-scale annotations therein, to nighttime scenes via the bridge of twilight time -- the time between dawn and sunrise, or between sunset and dusk. The goal of the method is to alleviate the cost of human annotation for nighttime images by transferring knowledge from standard daytime conditions. In addition to the method, a new dataset of road scenes is compiled; it consists of 35,000 images ranging from daytime to twilight time and to nighttime. Also, a subset of the nighttime images are densely annotated for method evaluation. Our experiments show that our method is effective for model adaptation from daytime scenes to nighttime scenes, without using extra human annotation.